A Method of Bearing Fault Feature Extraction Based on Improved Wavelet Packet and Hilbert Analysis

In order to supply a gap of current resonance vibration and STFT demodulation method applied to rolling bearing fault feature extraction of city rail vehicle, a fault diagnosis method for rolling bearing is presented, which is based on the integration of improved wavelet packet, frequency energy analysis and Hilbert marginal spectrum. When faults occur in rolling bearing, the energy of the rolling bearing vibration signal would change correspondingly, while the Hilbert energy spectrum can exactly provide the energy distribution of the signal in certain frequency with the change of the time and frequency. Thus, the fault information of the rolling bearing can be extracted effectively from the improved wavelet packet and Hilbert energy spectrum. The experimental result proves that the fault characteristic extracted by improved wavelet packet and Hilbert transform is in accord with the one analyzed from theory, and the fault feature extraction method is effective. The research results provide the theoretical foundation for the extraction of fault feature in rotary machine and have important practical value.